Service Quality Analysis of Online Travel Agencies (Ota) Using Multiclass Classification

Abstract

The simplicity provided by Online Travel Agencies (OTA) does not always make customers feel satisfied. Sometimes the customers get some problems with the company services. This finally led customers to give their opinion on social media. Large numbers of data in social media are capable to be an information source for the company to get customer insight. This study aims to determine the quality of OTA services based on customer opinions on social media Twitter. The method used in this study is a multiclass classification with Naïve Bayes Classifier model. Furthermore, each opinion is classified into positive and negative sentiment groups. Multiclass classification results show that Traveloka’s service quality is not good enough because six of the seven dimensions of service quality tend to have a negative sentiment. While the quality of Tiket.com and Pegipegi services can be assumed to be quite good because three of the seven dimensions of service quality get the more positive sentiment.